© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 69075
    Prev
    Next
    INDEX
    Explanations

    X&Y conjunctions

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron detects the ampersand when it appears in “R & D” (and similar “R & …” research-style acronyms).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Some
    -2.59
     Then
    -2.56
     There
    -2.48
     contaba
    -2.42
     if
    -2.38
     firmado
    -2.27
     ofrecemos
    -2.22
     q
    -2.16
     Selain
    -2.16
     Kemudian
    -2.14
    POSITIVE LOGITS
    amp
    2.64
    ین
    2.61
    >
    2.59
    E
    2.53
    t
    2.52
     citroen
    2.47
     incrí
    2.44
    v
    2.42
     rosario
    2.41
     vogel
    2.41
    Activations Density 0.009%

    No Known Activations